System and method for machine learning-temperature forecasting for automated tiering within a cloud storage system

Abstract

A method, computer program product, and computing system for forecasting a future temperature for a storage object within a multi-tiered cloud storage system. A cost associated with modifying a tiering of the storage object within the multi-tiered cloud storage system is determined based upon, at least in part, the future temperature forecasted for the storage object. The storage object is tiered in the multi-tiered cloud storage system based upon, at least in part, the cost associated with modifying the tiering of the storage object and a tiering policy associated with the multi-tiered cloud storage system.

What is claimed is: 
     
       1. A computer-implemented method, executed on a computing device, comprising:
 forecasting a future temperature for a storage object within a multi-tiered cloud storage system; 
 determining a cost associated with modifying a tiering of the storage object within the multi-tiered cloud storage system based upon, at least in part, the future temperature forecasted for the storage object; 
 tiering the storage object in the multi-tiered cloud storage system based upon, at least in part, the cost associated with modifying the tiering of the storage object and a tiering policy associated with the multi-tiered cloud storage system; 
 determining an actual temperature for the storage object for a predefined period of time; 
 determining an actual cost associated with tiering the storage object in the multi-tiered cloud storage system for the predefined period of time; and 
 adjusting one or more of an unsupervised machine learning model and the tiering policy based upon, at least in part, the actual temperature for the storage object for the predefined period of time, the future temperature for the storage object for the predefined period of time, and the actual costs associated with tiering the storage object for the predefined period of time. 
 
     
     
       2. The computer-implemented method of  claim 1 , wherein forecasting the future temperature for the storage object includes forecasting the future temperature using a time-series model and a plurality of previous temperature values for the storage object. 
     
     
       3. The computer-implemented method of  claim 1 , wherein forecasting the future temperature for the storage object includes forecasting the future temperature using the unsupervised machine learning model processing a plurality of input/output (IO) requests associated with the storage object. 
     
     
       4. The computer-implemented method of  claim 3 , wherein forecasting the future temperature using the unsupervised machine learning model includes generating a plurality of IO features using the plurality of IO requests associated with the storage object. 
     
     
       5. The computer-implemented method of  claim 4 , wherein the plurality of IO features include one or more of:
 an average amount of time between consecutive IO requests during a time interval; 
 a total amount of IO requests during the time interval; 
 a total amount of bandwidth during the time interval; 
 an average IO request size during the time interval; 
 an average amount of time between consecutive read IO requests during the time interval; 
 a frequency of activity during the time interval; and 
 an average amount of time between active time intervals of the plurality of time intervals. 
 
     
     
       6. The computer-implemented method of  claim 1 , wherein determining the cost associated with modifying the tiering of the storage object includes one or more of:
 determining a cost associated with promoting the storage object from a current cloud storage tier to a higher performance cloud storage tier within the multi-tiered cloud storage system; and 
 determining a cost associated with demoting the storage object from the current cloud storage tier to a lower performance cloud storage tier within the multi-tiered cloud storage system. 
 
     
     
       7. A non-transitory computer readable medium comprising a computer program product having a plurality of instructions stored thereon which, when executed by a processor, cause the processor to perform operations comprising:
 forecasting a future temperature for a storage object within a multi-tiered cloud storage system; 
 determining a cost associated with modifying a first tiering of the storage object within the multi-tiered cloud storage system based upon, at least in part, the future temperature forecasted for the storage object; 
 tiering the storage object in the multi-tiered cloud storage system based upon, at least in part, the cost associated with modifying the first tiering of the storage object and a tiering policy associated with the multi-tiered cloud storage system; 
 determining an actual temperature for the storage object for a predefined period of time: 
 determining an actual cost associated with tiering the storage object in the multi-tiered cloud storage system for the predefined period of time; and 
 adjusting one or more of an unsupervised machine learning model and the tiering policy based upon, at least in part, the actual temperature for the storage object for the predefined period of time, the future temperature for the storage object for the predefined period of time, and the actual costs associated with tiering the storage object for the predefined period of time. 
 
     
     
       8. The computer program product of  claim 7 , wherein forecasting the future temperature for the storage object includes forecasting the future temperature using a time-series model and a plurality of previous temperature values for the storage object. 
     
     
       9. The computer program product of  claim 7 , wherein forecasting the future temperature for the storage object includes forecasting the future temperature using the unsupervised machine learning model processing a plurality of input/output (IO) requests associated with the storage object. 
     
     
       10. The computer program product of  claim 9 , wherein forecasting the future temperature using the unsupervised machine learning model includes generating a plurality of IO features using the plurality of IO requests associated with the storage object. 
     
     
       11. The computer program product of  claim 10 , wherein the plurality of IO features include one or more of:
 an average amount of time between consecutive IO requests during a time interval; 
 a total amount of IO requests during the time interval; 
 a total amount of bandwidth during the time interval; 
 an average IO request size during the time interval; 
 an average amount of time between consecutive read IO requests during the time interval; 
 a frequency of activity during the time interval; and 
 an average amount of time between active time intervals of the plurality of time intervals. 
 
     
     
       12. The computer program product of  claim 7 , wherein determining the cost associated with modifying the tiering of the storage object includes one or more of:
 determining a cost associated with promoting the storage object from a current cloud storage tier to a higher performance cloud storage tier within the multi-tiered cloud storage system; and 
 determining a cost associated with demoting the storage object from the current cloud storage tier to a lower performance cloud storage tier within the multi-tiered cloud storage system. 
 
     
     
       13. A computing system comprising:
 a memory; and 
 a processor configured to forecast a future temperature for a storage object within a multi-tiered cloud storage system, to determine a cost associated with modifying a first tiering of the storage object within the multi-tiered cloud storage system based upon, at least in part, the future temperature forecasted for the storage object, to tier the storage object in the multi-tiered cloud storage system based upon, at least in part, the cost associated with modifying the first tiering of the storage object and a tiering policy associated with the multi-tiered cloud storage system, to determine an actual temperature for the storage object for a predefined period of time, to determine an actual cost associated with tiering the storage object in the multi-tiered cloud storage system for the predefined period of time, and to adjust one or more of an unsupervised machine learning model and the tiering policy based upon, at least in part, the actual temperature for the storage object for the predefined period of time, the future temperature for the storage object for the predefined period of time, and the actual costs associated with tiering the storage object for the predefined period of time. 
 
     
     
       14. The computing system of  claim 13 , wherein forecasting the future temperature for the storage object includes forecasting the future temperature using a time-series model and a plurality of previous temperature values for the storage object. 
     
     
       15. The computing system of  claim 13 , wherein forecasting the future temperature for the storage object includes forecasting the future temperature using the unsupervised machine learning model processing a plurality of input/output (IO) requests associated with the storage object. 
     
     
       16. The computing system of  claim 15 , wherein forecasting the future temperature using the unsupervised machine learning model includes generating a plurality of IO features using the plurality of IO requests associated with the storage object. 
     
     
       17. The computing system of  claim 16 , wherein the plurality of IO features include one or more of:
 an average amount of time between consecutive IO requests during a time interval; 
 a total amount of IO requests during the time interval; 
 a total amount of bandwidth during the time interval; 
 an average IO request size during the time interval; 
 an average amount of time between consecutive read IO requests during the time interval; 
 a frequency of activity during the time interval; and 
 an average amount of time between active time intervals of the plurality of time intervals. 
 
     
     
       18. The computing system of  claim 13 , wherein determining the cost associated with modifying the tiering of the storage object includes one or more of:
 determining a cost associated with promoting the storage object from a current cloud storage tier to a higher performance cloud storage tier within the multi-tiered cloud storage system; and 
 determining a cost associated with demoting the storage object from the current cloud storage tier to a lower performance cloud storage tier within the multi-tiered cloud storage system.